Patient Care

11 Things Gastroenterologists Think Hospitalists Need to Know


11 Things Gastroenterologists Think Hospitalists Need to Know


So many symptoms that are staples of gastroenterology—chest pain, nausea, diarrhea—are mainstay causes for hospitalization that it might be worth fine-tuning how well you handle patients with gastroenterology disorders.

The Hospitalist asked several gastroenterologists for their guidance on better care and their suggestions for correcting some common mistakes that they encounter. Here are their tips:

1 Fluid resuscitation is crucial for pancreatitis patients.

It’s very important to rehydrate these patients within the first 24 hours, because those who remain underhydrated can have a worse prognosis, says Robert Coben, MD, academic coordinator for the Gastroentestinal Fellowship Program at Thomas Jefferson University Hospital in Philadelphia. On occasion, physicians are reluctant to give extra fluids to these patients, he says, particularly if they have heart failure or suffer kidney problems. A 70-kg patient should be receiving about 200 cc per hour, he notes.

“Sometimes we’ll walk in the room and they’re getting 80 cc an hour,” Dr. Coben says.

“These patients…need to be flooded with fluids,” says Rajeev Jain, MD, chief of gastroenterology at Presbyterian Hospital of Dallas, partner at Texas Digestive Disease Consultants, and chair of the Practice Management and Economics Committee of the American Gastroenterological Association. “We’re talking sometimes liters and liters of IV normal saline or lactated Ringer’s (solution) in a 24-hour period.”1

Marcelo Vela, MD, a gastroenterologist and hepatologist at the Mayo Clinic in Scottsdale, Ariz., and an associate editor with Clinical Gastroenterology and Hepatology, says Ringer’s solution is a better choice than normal saline.2

“If you’re going to start IV fluids on somebody who’s coming in with acute pancreatitis, Ringer’s solution has been shown to be superior to saline in randomized controlled trials,” Dr. Vela says. “It reduces systemic inflammation.”

2 Gastrointestinal bleeding decisions

When inpatients have gastrointestinal bleeding, the hospitalist often has to assess its severity and make the call on whether a patient needs the ICU.

“The most important thing for that is obviously the vital signs,” says John Pandolfino, MD, chief of the division of medicine–gastroenterology and hepatology at Northwestern University Feinberg School of Medicine in Chicago. “If people are tachycardic and they’re not responding to hydration and blood transfusion, that usually means it’s a pretty active bleed and they need to go to the intensive care unit. If you have somebody who’s GI bleeding and they’re coagulopathic (i.e., they’re on anti-coagulation because they have a valve and they need anti-coagulation or they have cancer or bad cardiovascular disease), those are the people that you should have a low threshold for sending to the intensive care unit with a GI bleed; those are the people who are at a very high mortality [risk].”

He added that those with an ulcer, with a visible vessel, are at a high-risk of a rebleed and should spend some time in the ICU.

“Those people should be evaluated in the intensive care unit for at least 24 hours, maybe even 72 hours,” Dr. Pandolfino says, “and they should have IV PPI [proton pump inhibitor] therapy.”

3 Endoscopy has very low yield for diagnosis of reflux.

“Endoscopy has good yield for mucosa abnormalities on inspection of the esophageal mucosa, but it does not give you a diagnosis of reflux, especially on patients who have already been treated with a PPI,” says Prakash Gyawali, MD, MRCP, professor of gastroenterology at Washington University in St. Louis, Mo. “So, usually, in those settings, obtaining a consult or trying to decide exactly what you’re looking for to explain the symptoms has better yield than an endoscopy.”

Sometimes a pH study is needed, but that has to be planned, because patients may have to be taken off of a PPI in advance. That means those studies are not easy to set up in the hospital and might best be arranged by the gastroenterologist, Dr. Gyawali says.

4 In cases of acute diarrhea, order a stool sample right away.

That will help guide care from the gastroenterologist, if and when the gastroenterologist is called in, Dr. Pandolfino says.

“One of the things that is frustrating for the gastroenterologist is that we get called initially, but really the hospitalist should be getting stool studies, and they should have at least a very good idea of what they need from us,” he says. “Because, really, endoscopy is not usually needed very often in diagnosis of acute diarrhea.”

Broad-range stool studies, a good history and physical, and examining labs for a possible chronic inflammatory process or anemia are good ways to begin to assess patients with diarrhea, he says. Clostridium difficile colitis has to be considered as well, Dr. Pandolfino says.

Endoscopy is more helpful in evaluating acute diarrhea in those with bloody diarrhea suspected of having inflammatory bowel disease or an infectious diarrhea but on whom cultures have come back negative. For those with compromised immune systems, endoscopy could be done earlier, as well.

“So for us, I think we really need to get into the picture a little bit after the patient has been brought into the hospital and the stool study is negative, unless they’re an immune-compromised patient,” Dr. Pandolfino says.

5 When—and how—to test the stool.

If a patient develops diarrhea while already in the hospital, the only stool test needed is C. diff.

“They shouldn’t be developing a viral diarrhea, they shouldn’t be developing an infectious diarrhea—let’s say, from E. coli or Salmonella—unless they literally developed it a couple hours after getting in,” Dr. Jain says. “It should either be C. diff or a side effect of some medication. … We don’t need to spend the extra money, which is of low-value care to send for OVA and parasites, or bacterial pathogens and so forth.”

Dr. Jain says he thinks such testing is being done more appropriately of late.

“But I still will see multiple stool tests sent on somebody who’s been in the hospital for a week and then develops diarrhea,” he explains.

6 Gastroenterologists do not need to be consulted for every C. diff infection.

“I think that we really should get involved when patients are either not responding or when they’re very ill,” Dr. Pandolfino says.

Hospitalists should consider whether patients are on antibiotics or a PPI and whether or not they need to stay on those medications. Also, medications that slow motility (i.e., loperamide) should be avoided, if possible.

“We don’t want it to linger,” Dr. Pandolfino says. “One of the basic mechanisms of how we get rid of pathogens is to expel them with diarrhea.

“But you certainly don’t want the patient to be uncomfortable to the point where they’re having 20 to 30 bowel movements a day.”

In non-severe patients who have some diarrhea, abdominal pain, nausea, and vomiting—but are able to keep food down—the specialists might not have to be called in, and patients can just be treated with oral metronidazole or oral vancomycin fairly simply, he says.

For those who seem severely ill with a dilating colon, are in nearly a septic state, and have very severe diarrhea, the gastroenterologist should probably be called in, he says.

“Even if they have underlying heart failure, if they’re volume-depleted, they need that volume. Sometimes you’ll see that it takes eight hours to get two units of blood in. That’s inadequate.” —Rajeev Jain, MD

7 For patients with a possible GI bleed and black stools, do an exam before calling in the gastroenterologist.

The exam should determine what the stool color is and whether it is heme-positive, and the patient’s blood count should be evaluated, Dr. Coben says.

Frequently, “the consultant’s the one who ends up doing the rectal exam and checking that,” he says. “Sometimes we find that [the patient is] really not bleeding,” and it was just a case of a hospitalist taking the patient’s word that they were bleeding.

Being on iron therapy or taking Pepto-Bismol can turn the stool black, and the stool might not really be black; colors can sometimes be open to interpretation.

But he cautions that colon cancer could be the cause for a GI bleed.

“We’ve had it happen a few times where this occurred,” he says. “The patient was discharged, and they really didn’t get proper follow-up, and it ended up that they had a colon cancer. It kind of delayed that diagnosis. So I think you have to be aware of [the fact that], especially in somebody over the age of 40 or 50, if they have an iron-deficiency, anemia, or heme-positive stool, the first thing you really need to exclude is a colon cancer.”

8 Minimize CT scans in early evaluation and management of acute pancreatitis patients.

“The reason for that is they tend to be intravascularly volume-depleted,” Dr. Jain says. “So [with] the IV contrast, there’s an increased risk of developing kidney failure. It’s also been associated with increased risk of necrotizing pancreatitis.”

Dr. Jain notes that when he is consulted on this kind of patient, he will order a sonogram. If it doesn’t show gallstones, and there’s no clear reason for the pancreatitis, he will want a CT scan, but he will wait a few days until the patient is fluid-resuscitated.

Dr. Jain says that this is a problem more often seen in the ED—where 90% of patients with acute pancreatitis have already gotten the CT scan—and less so among hospitalists, but it’s still worth the reminder.

A CT scan right away is justified to rule out something such as a perforation, but not in the case of classic symptoms of acute pancreatitis, he adds.

9 Actively bleeding patients?

It’s smart to give patients a unit of packed red blood cells more quickly if they’re actively bleeding. Even over just one hour is OK, Dr. Jain says.

“Even if they have underlying heart failure, if they’re volume-depleted, they need that volume,” Dr. Jain explains. “Sometimes you’ll see that it takes eight hours to get two units of blood in. That’s inadequate.”

Another point worth a reminder, he says: Two large-bore peripheral IV’s are “much, much better” than a central or PICC line to deliver IV fluid resuscitation.

Prakash Gyawali, MD, MRCP

Image credit: Courtesy of Washington University School of Medicine

“Sometimes hospitalists order barium studies because these can be done the same day, whereas with endoscopy, patients need to be put on a schedule and have to be NPO [nothing by mouth] overnight.” —Prakash Gyawali, MD, MRCP

10 Don’t be too quick to order barium studies, especially in patients with dysphagia.

“The problem with that is it takes much longer to do an endoscopy if barium is put in the esophagus, because we usually wait until the barium clears,” Dr. Gyawali says. “And inpatient evaluation of new-onset dysphagia should be endoscopy first, and not barium, because biopsies need to be taken to rule out eosinophilic esophagitis.

“Sometimes hospitalists order barium studies because these can be done the same day, whereas with endoscopy, patients need to be put on a schedule and have to be NPO [nothing by mouth] overnight.”

11 Gastric-emptying studies should be outpatient.

Gastric-emptying studies are often better done when patients are not in a hospital, experts say, because they might be on medications that would interfere with the study.

“A common issue, not just among hospitalists but also gastroenterologists, is that patients may be on a bunch of medicines that would affect stomach-emptying while they are in-house for some other problem,” Dr. Gyawali says. “A lot of times, these patients with pain may get narcotics. They may be on medicines to prevent them from throwing up. … And all of these will slow down gastric emptying.”

With an abnormal test result, “you are left to decide whether that is a real abnormality or whether the medicines the patients were on impacted the abnormality.”

If emptying was significantly prolonged, the test may have some value, “but then it probably will need to be corroborated with symptoms and with endoscopic findings.”

Thomas Collins is a freelance writer in South Florida.


  1. Warndorf MG, Kurtzman JT, Bartel MJ, et al. Early fluid resuscitation reduces morbidity among patients with acute pancreatitis. Clin Gastroenterol Hepatol. 2011;9(8):705-709.
  2. Wu BU, Hwang JQ, Gardner TH, et al. Lactated Ringer’s solution reduces systemic inflammation compared with saline in patients with acute pancreatitis. Clin Gastroenterol Hepatol. 2011;9(8):710-717.

Next Article:

   Comments ()